Talk:Paired difference test

Welch's test not relevant
I believe that the Welch's test and homogeneity of variance should not be mentioned in this article: when using a paired test, we effectively turn the question into a single sample test for the differences. See also http://yatani.jp/HCIstats/TTest#PairedUnpaired for a claim that paired t-test does not assume homogeneity. Am I correct? Anša (talk) 16:39, 9 December 2013 (UTC)


 * That is exactly right. A paired t-test does not require equal population variance in two samples, since there is effectively only one sample. 172.91.107.147 (talk) 05:41, 28 March 2024 (UTC)

External links modified
Hello fellow Wikipedians,

I have just modified one external link on Paired difference test. Please take a moment to review my edit. If you have any questions, or need the bot to ignore the links, or the page altogether, please visit this simple FaQ for additional information. I made the following changes:
 * Added tag to http://compbio.berkeley.edu/people/ed/SeqCompEval/Pairwise
 * Replaced archive link http://webcache.googleusercontent.com/search?q=cache:UWWnuePMeTgJ:compbio.berkeley.edu/people/ed/SeqCompEval/+http://compbio.berkeley.edu/people/ed/SeqCompEval/Pairwise&cd=1&hl=en&ct=clnk&source=www.google.com with https://web.archive.org/web/20080629033730/http://compbio.berkeley.edu/people/ed/SeqCompEval/ on http://compbio.berkeley.edu/people/ed/SeqCompEval/

When you have finished reviewing my changes, you may follow the instructions on the template below to fix any issues with the URLs.

EDIT: First link (extreme values) is dead. Should be removed?

Cheers.— InternetArchiveBot  (Report bug) 23:13, 23 May 2017 (UTC)